skills/vercel-labs/emulate/aws/Gen Agent Trust Hub

aws

Pass

Audited by Gen Agent Trust Hub on Apr 18, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • Command Execution for Service Interaction: The skill utilizes curl and npx commands to interact with a local server. These operations are intended for managing the emulator and executing test queries against it within a local environment.
  • External Package Retrieval: The skill documentation instructs users to execute npx emulate, which downloads and runs a package from an external registry. This is the primary method for initializing the AWS emulation service.
  • Use of Recognized Example Credentials: The configuration examples employ standard AWS documentation placeholder keys (e.g., AKIAIOSFODNN7EXAMPLE). These are widely used for testing purposes and are not actual sensitive credentials.
  • Indirect Prompt Injection Surface: The skill handles data that could be influenced by external sources, such as S3 object content or SQS messages, creating a potential surface for indirect instructions.
  • Ingestion points: Content stored in S3 buckets, SQS message bodies, and IAM policy definitions (SKILL.md).
  • Boundary markers: The provided examples do not use specific delimiters or warnings to isolate processed data from agent instructions.
  • Capability inventory: The skill environment allows for network interaction via curl and package execution via npx (SKILL.md).
  • Sanitization: No explicit data sanitization or validation steps are demonstrated in the example workflows.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 18, 2026, 07:59 AM